Skillv1.2.2

ClawScan security

Stock Monitor · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 7, 2026, 12:00 PM

Verdict: benign
Confidence: medium
Model: gpt-5-mini
Summary: The skill's code, instructions, and requested resources are coherent with a stock-monitoring/reporting tool; it reads local stock/position JSON files and fetches market data from public finance APIs and does not request unrelated credentials — but review the full sources and Feishu/cron push configuration before enabling automated runs.
Guidance: This skill appears to do what it claims: run local Python scripts to fetch market data, compute indicators, read ~/.openclaw JSON files (stock-pool, positions, trades, alerts), and push reports to a Feishu group via the platform's messaging tool. Before installing or enabling automated runs: 1) Review the full contents of the provided scripts (some snippets were truncated in the listing) to ensure no unexpected network endpoints or data exfiltration code exist. 2) Audit the ~/.openclaw JSON files — they contain sensitive financial/position data — and consider file permissions or using a separate account/VM for automation. 3) Confirm how report delivery to Feishu is configured on your OpenClaw platform (check that no secret tokens need to be added or that they are stored/rotated safely). 4) If you will enable cron/autonomous runs, verify the cron job configuration (times, target Feishu group) and that rate limits for external APIs are acceptable. If you want, I can scan the full untruncated script files for any hidden or suspicious network calls or credential access.

Review Dimensions

Purpose & Capability: okName/description (stock monitoring, indicators, reports) matches the files and actions: Python scripts fetch market data (Tencent, EastMoney), compute indicators, read ~/.openclaw JSON config (stock pool, positions, trades), and generate/push reports. No unrelated services or credentials are requested.
Instruction Scope: noteSKILL.md and prompt templates explicitly instruct reading ~/.openclaw/*.json files, running the included scripts, fetching web news, and pushing reports to Feishu. These actions are within the declared purpose. Note: the skill will read local holdings/transaction files (sensitive financial data) and will perform network requests to finance APIs and news sites as part of normal operation.
Install Mechanism: okNo install spec is present (instruction + bundled scripts). This is lowest-risk from installer perspective — files are provided and executed by the agent; nothing is downloaded at install time by the skill itself.
Credentials: noteThe skill declares no required environment variables or external credentials. It does depend on platform-level messaging (Feishu) for report delivery; SKILL.md assumes a message/send tool and a Feishu group ID but does not request tokens. Confirm the platform provides the messaging integration or check whether you must supply Feishu credentials separately. It legitimately reads local config files under ~/.openclaw but those can contain private trading data, so access is proportional but sensitive.
Persistence & Privilege: okalways is false and the skill is user-invocable; it defines cron schedules and IM delivery but does not request persistent elevated privileges or modify other skills. Autonomous invocation is allowed by default (normal).